Accounting for 9% to 12% of all shoulder injuries and 3.2% of all joint dislocations, acromioclavicular joint (ACJ) dislocation is a common injury. There are many surgical treatments for ACJ dislocation, among which the reconstruction of coracoclavicular ligament has been accepted by the majority of orthopedic doctor. However, pain and loss of function are results of all kinds of surgery, especially for those heavy manual workers. Therefore, the gold standard treatment for ACJ dislocation still remains controversial.

1. Research objects: From 2009 to 2016, we treated 62 patients (54 males and 8 females) with ACJ dislocation via single and double bundle coracoclavicular ligament reconstruction. The mean age was (53±8.1) years. 28 cases had the right side affected, and 34 cases had the left side affected. There were 29 cases of fall, 20 cases of car accident, 6 cases of sports injury, 3 cases of fall injury at high place and 4 cases of other injuries. The condylar fracture and mutation were excluded. Rockwood classification: 35 cases of degree III, 8 cases of degree IV, 19 cases of degree V; Injury to operation time: 59 cases within 2 weeks, 3 cases over 3 weeks; Surgical method: 51 cases of single bundle reconstruction of coracoclavicular ligament (Figure 1) , 11 cases of double bundle reconstruction (Figure 2) . 48 patients were reconstructed with single bundle Endobutton, and 14 patients were reconstructed with Arthrex AC Joint TightRope (3 single bundle cases and 11 double bundle cases) . 2. Surgical approach: The patient was put in supine position or beach chair position under nerve block or general anesthesia. A 3-4 cm incision was cut along the outer end of clavicle. The acromioclavicular joint was revealed, and the broken joint disc was cleaned. Then, the reduction of acromioclavicular joint was achieved. The reduction criterion was that the leading edge of outer clavicle reached the end of coracoacromial ligament, and the outer edge was pressed down to the inner side of acromion without gap. A 1.5 mm Kirshner wire was used to fix the acromioclavicular joint from front to back percutaneously (avoiding the tunnel) ; then, a 2 cm vertical incision was made on condyle, entering the medial side of conjoint tendon and reaching the flat surface under condyle by finger. With finger as guidance, the front of the flat surface or the posterior portion of condyle was aimed. For the single bundle group, a 2.0 mm Kirschner wire was used to perforate the condyle from the midpoint of the 35 mm outer end of clavicle, and the 4.0 mm diameter drill was used for drilling. For the double bundle group, a 2.0 mm Kirschner wire was used to perforate the condyle from the midpoint of the 40 mm outer end of clavicle, and the 4.0 mm diameter drill was used for drilling. Later, a drilling was made at the midpoint of the 20 mm outer end of clavicle with 4.0 mm drill. The clavicle had two channels, and the condyle had one single channel. A double-strand steel wire of 0.8 mm was taken out from the clavicle side to the condyle side through the channel. Then, the titanium plate for each group was fixed. Finally, the condyle side wire was fixed on conjoint tendon surface, and the acromioclavicular joint capsule was repaired and strengthened. As the Kirschner wire was buried in soft tissue, the incision was sutured. 3. Postoperative treatment: The limb was suspended for 3 to 4 weeks with triangle towel postoperatively. On the second postoperative day, the active function rehabilitation of the fingers, wrists and elbow joints were performed. On the third day, the passive shoulder joint pendulum exercises could be started. The digital radiography (DR) of 2, 3, 6 months postoperatively were observed to evaluate the acromioclavicular joint reduction. The Kircher wire was removed 6 to 8 weeks after surgery. limb support and lifting weight were prohibit within 2 months. Independent activities were allowed after 3 months. 4. Efficacy evaluation: The Karlsson and Constant-Murley criteria were used to evaluate the recovery of shoulder function in patients 1 year after surgery. 5. Statistical analysis: Statistical analysis was performed using SPSS 13.0 software. The measurement data were expressed as±s, and the paired t test was used. P<0.05 was considered statistically significant.

All the incisions healed during the first stage, and there was no internal fixation breakage, shedding or failure. No complication such as iatrogenic condyle/clavicular fracture or vascular/nerve injury occurred. All patients were followed up for 1-7 years with an average of 2.3 years. According to the Karlsson evaluation criteria, there were 57 cases of excellence, 1 case of good and 3 cases of acceptable, and the excellent and good rate was 95% 1 year after the operation. The total and section scores of Constant-Murley shoulder function were significantly better than those before the surgery (P mean < 0.01) . There were two cases of internal fixation failure. In one case, the condyle channel was deviated from the midpoint, which caused the rupture of lateral wall. The other case was caused by poor patient compliance. There were two cases of acromioclavicular joint osteoarthritis, and all were operated with single-bundle reconstruction. Postoperative stress bone resorption and osteolysis were common in the clavicular channel, the single-bundle reconstruction group and the patients with large diameter and osteoporosis. Two cases of heterotopic ossification occurred around the reconstructed coracoclavicular ligament.

ACJ dislocation is a common disease in orthopedics, and the postoperative complications are receiving increasing amount of concern. The clinical research has been carried out in terms of anatomy and biomechanics. In this study, we used single and double bundle coracoclavicular ligament reconstructions to treat ACJ dislocation. The operation was simple and minimally invasive, and the clinical results were satisfactory.